The wireless LAN is gathering attention as a system to free users from LAN cabling. With a wireless LAN, the greater part of cables can be omitted from the workspace such as offices and the like, so communication terminals such as personal computers (PCs) can be moved with relative ease. In recent years, increased speed and reduced process of wireless LAN systems has led to marked increase in demand thereof. Particularly, as of recent, introduction of the personal area network (PAN), wherein a small-scale wireless network is configured with multiple electronic devices which people have nearby and communication of information is performed, is being considered. Differing wireless communication systems and wireless communication devices have been stipulated, using frequency bandwidths which do not require licensing by the regulatory authorities, such as the 2.4 GHz band and the 5 GHz band, for example.
An example of a commonplace standard relating to wireless networks is IEEE (The Institute of Electrical and Electronic Engineers) 802.11 (see Non-patent Document 1, for example), Hiper LAN/2 (see Non-patent Document 2 or Non-patent Document 3, for example), IEEE 802.15.3, Bluetooth communication, and so forth. Under the IEEE 802.11 Standard, various wireless communication methods exist according to the communication method and frequency band used, such as the IEEE 802.11a Standard, IEEE 802.11b Standard, and so on.
In order to configure a local area network using wireless techniques, with a commonly-used method, a device serving as a control station, that is called an “access point” or “coordinator” is set up within the area, and a network is formed under the centralized control of this control station.
With a wireless network in which an access point has been set up, an access control method based on band reservation is widely employed wherein, in the event of transmission of information from a certain communication device, first, the band necessary for transmission of that information is reserved at the access point so that transmission path usage is implemented such that no collision with information transmission from other communication devices. That is to say, setting up the access point allows synchronous wireless communication wherein communication devices within the wireless network are synchronized with each other.
However, there is the problem with a wireless communication system which has an access point in that asynchronous communication between transmitting and receiving communication devices always necessitates wireless communication via the access point, meaning that the usage efficiency of the transmission path is halved.
As opposed to this, “Ad-hoc communication”, wherein communication terminals directly perform asynchronous wireless communication, is being proposed as another method for configuring a wireless network. For small-scale wireless networks configured of a relatively small number of clients which are in the proximity of each other in particular, ad-hoc communication wherein arbitrary terminals can directly perform asynchronous wireless communication without using a predetermined access point is thought to be appropriate.
A central control station does not exist in an ad-hoc wireless communication system, and accordingly is suitable for configuring a home network made up of home electronic appliances, for example. Features of an ad-hoc network are that the network does not readily fail since even in the event that one device malfunctions or the power thereof is turned off, the routing is automatically changed, data can be transmitted over relatively long distances while maintaining a high-speed data rate by hopping packets multiple times between mobile stations, and so forth. Various development examples of ad-hoc systems are known (see Non-patent Document 4, for example).
For example, IEEE 802.11 wireless LAN systems have an ad-hoc mode for operating peer-to-peer in an autonomous distributed manner without a control station begin provided. Under this operating mode, at the beacon transmission time each terminal counts a random period, and in the event that the device has not received a beacon from another terminal by the time that the period ends, transmits its own beacon.
Now, conventional wireless networking will be described in detail with reference to the example of IEEE 802.11.
Networking with IEEE 802.11 is based on the concept of BSS (Basic Service Set). There are two types of BSS; one being a BSS defined by an infrastructure mode wherein a master such as an AP (Access Point: control station) exists, and an IBSS (Independent BSS) defined by an ad-hoc mode configured only of multiple MTs (Mobile Terminals).
Infrastructure Mode
The operations of IEEE 802.11 when in the infrastructure mode will be described with reference to FIG. 23. With an infrastructure mode BSS, an AP to perform coordination within the wireless communication system is indispensable.
The AP handles the range where the airwaves reach around itself as a BSS, thus configuring a “cell” as it is called in a so-called cellular system. An MT nearby the AP is contained in the AP, and participates in the network as a member of the BSS. That is to say, the AP transmits control signals called beacons at appropriate time intervals, the MT capable of receiving these beacons recognizes that an AP is nearby, and further, a connection is established with the AP.
With the example shown in FIG. 23, the communication station the STA0 operates as the AP, and the other communication stations the STA1 and the STA2 operate as MTs. Now, as indicated in the chart to the right side of the drawing, the communication station the STA0 serving as the AP transmits beacons a predetermined time intervals. The transmission point-in-time of the next beacon is notified within the beacon by a parameter format known as Target Beacon Transmit Time (TBTT). Upon the time of the TBTT coming, the AP operates beacon transmission procedures.
Conversely, by receiving the beacon, MTs nearby the AP can recognize the next beacon transmission time by decoding of the internal TBTT field, so in some cases (in cases wherein there is no need for reception), the receiving device may turn off the power and go to sleep until the next or until several TBTTs in the future.
In the infrastructure mode, only the AP transmits beacons a predetermined frame cycles. On the other hand, the nearby MTs succeed in participating in the network by receiving the beacons form the AP, and do not transmit beacons themselves. Note that the focus of the present invention is to operate a network without a master control station such as an AP being directly involved, so the infrastructure mode will be discussed no further.
Ad-Hoc Mode
The operations of IEEE 802.11 when in the other ad-hoc mode will be described with reference to FIG. 24.
With IBSS in the ad-hoc mode, multiple MTs perform negotiation one with another, and subsequently autonomously define the IBSS. Upon defining the IBSS, at the end of the negotiation the MT group determines the TBTT every predetermined interval. Upon recognizing that the TBTT has arrived by referencing a clock within itself, following a random time delay each MT transmits a beacon in the event of recognizing that no one has transmitted a beacon yet.
In the example shown in FIG. 24, the manner in which two MTs make up an IBBS is illustrated. In this case, one of the MTs belonging to the IBSS transmits a beacon each time the TBTT arrives. This also includes cases wherein beacons transmitted from the MTs collide.
There are also cases with IBSS wherein the MTs turn off the power of the transmitting/receiving device and go to sleep as necessary. However, the sleep state is not directly related to the essence of the present invention, and accordingly description thereof will be omitted in the present specification.
Transmission/Reception Procedures Under IEEE 802.11
Next, transmission/reception procedures under IEEE 802.11 will be described.
It is know that with wireless LAN networks under an ad-hoc environment, a hidden terminal problem generally occurs. A hidden terminal is a terminal which can be heard from one communication station which is the other part of communication therewith in a case of communication being carried out between certain communication stations, but cannot be heard by other communication stations, and since negotiation cannot be performed between hidden terminals, there is the possibility that transmission operations may collide.
CSMA/CA according to RTS/CTS procedures is known as a methodology for solving the hidden terminal problem. IEEE 802.11 also employs this methodology.
Now, CSMA (Carrier Sense Multiple Access with Collision Avoidance: Carrier Sense Multiple Access) is a connection method for performing multiple access based on carrier detection. Since receiving signals of information transmitted from a local station is difficult in wireless communication, collision is avoided by starting information transmission from the local station following confirming that there is no information transmission from other communication devices with the CSMA/CA (Collision Avoidance) method rather than CSMA/CD (Collision Detection). The CSMA method is an access method suitable for asynchronous data communication such as file transfer and electronic mail.
With the RTS/CTS method, transmission of data is started in response to a communication station which is the data originator transmitting a transmission request packet RTS (Request To Send), and a confirmation notification packet CTS (Clear To Send) being received form the communication station which is the data transmission destination. Upon a hidden terminal receiving at least one of an RTS or CTS, a transmission stop period is set of the local station for a period during which data transmission based on RTS/CTS procedures is predicted, whereby collision can be avoided.
FIG. 25 illustrates an operation example of RTS/CTS procedures. With the example shown in the drawing, an example is illustrated of a case wherein information (Data) is transmitted from a communication station the STA0 to a communication station the STA1 which mutually perform communication operations in an autonomously dispersed manner.
First, prior to actual information transmission, the STA0 confirms that the media is clear for a predetermined time, following which the RTS packet is transmitted to the STA1, which is the destination of the information, following CSMA procedures. In response o reception of the RTS packet, the STA1 transmits a CTS packet to the STA0 which gives feedback to the effect that the RTS has been received.
In the event that the CTS is successfully received, the STA0 which is the transmitting side determines that the media is clear, and promptly transmits the information (Data) packet. Also, upon successfully receiving the information, the STA1 returns an ACK, whereby one packet of RTS/CTS transmission/reception transaction ends.
In the event that another station has happened to have transmitted some sort of signal at the same time as the STA0 which is the information originator transmitting the RTS, the STA1 which is the information recipient cannot receive the RTS due to the signals colliding. In this case, the STA1 does not return a CTS. As a result, the STA0 can recognize that the earlier RTS has collided, since no CTS is received for a while. Then, procedures for resending the RTS with a random backoff are activated at the STA0. Basically, competition is carried out for wining transmission rights while bearing the risk of such collision.
Access Competition Method in IEEE 802.11
Next, the access competition method stipulated in IEEE 802.11 will be described.
With IEEE 802.11, four types of packet intervals (IFS: Inter Frame Space) are defined. Here, three of these IFSs will be described with reference to FIG. 26. The IFSs defined are, in order from the shorter, SIFS (Short IFS), PIFS (PCF IFS), and DIFS (DCF IFS).
With IEEE 802.11, CSMA is employed as a basic media access procedures (described above), however, it should be noted that a transmission right is granted to the transmitting device only in a case wherein a backoff timer is operated over a random time period while monitoring the media state before transmitting something, and there are no signals transmitted during this period.
In the case of transmitting normal packets following the CSMA procedures (DCF (called Distributed Coordination Function), following some sort of packet transmission ending, first, the media state is monitored by DIFS, and in the event that there are no transmission signals during this period, a random backoff is taken, and further, in the event that there are no transmission signals during this period as well, transmission rights are granted.
On the other hand, transmission of packets with extraordinarily high urgency, such as ACK, is permitted following SIFS packet intervals. This enables packets with high urgency to be transmitted before packets transmitted following normal CSMA procedures.
To summarize the above, the reason that differing types of packet interval IFSs are defined is that prioritizing is performed in the packet transmission competition, according to whether the IFS is SIFS, PIFS, of DIFS, i.e., according to the packet interval length. The purpose of using PIFS will be described later.
Band Guarantee (1) Under IEEE 802.11
In a case of access competition with CSMA, guaranteeing and securing a certain band is impossible. Accordingly, IEEE 802.11 has PCF (Point Coordination Function) to serve as a mechanism for guaranteeing and securing a band. However, PCF is based on poling, and does not operate under Ad-hoc but is only performed under management of an AP in the infrastructure mode.
FIG. 27 illustrates the way in which preferential communication is provided by PCF operations. In the drawing, the STA0 operates as an AP, and the STA1 and the STA2 participate in the BSS managed by the AP. This case assumes the STA1 transmitting information while guaranteeing band.
After transmitting a beacon for example, the STA0 serving as the AP performs poling by transmitting a CF-Poll message to the STA1 at a SIFS interval. The STA1 which has received the CF-Poll is granted data transmission rights, and transmission of data at the SIFTS interval is permitted, in response to this, the STA1 transmits data following SIFS. Upon the STA0 returning an ACK to the transmitted data, and one transaction ending, the STA0 polls the STA1 again.
In the example shown in FIG. 27, a case wherein this poling has failed for some reason is shown. At this time, upon recognizing that information is not transmitted from the STA1 following SIFS after poling the STA1 again, the STA0 deems the poling to have failed, and performs poling again following a PIFS interval. In the event that the poling retry succeeds, data is transmitted from the STA1, and an ACK is returned from the STA0.
Even in the event that the STA2 has a transmitted packet, for example, during this series of procedures, the transmission right never shifts to the STA2, since this would mean that the STA0 or the STA1 would be transmitting at a SIFS or PIFS interval before the DIFS time interval elapses. That is to say, the STA1, which has been polled by the STA0 serving as the AP, always has the transmission right.
Band Guarantee (2) Under IEEE 802.11
Further band guarantee means are being studied for IEEE 802.11, and implementation of a technique called Enhanced DCF (EDCF) is planned (the QoS enhancement in IEEE 802.11e).EDCF is arranged such that the width for which a random backoff value can be set is short for urgent traffic needing band guarantee, and the width for which the packet intervals IFS and backoff value shown in FIG. 26 can be set is longer for other traffic. Consequently, a mechanism is realized which enables transmission of urgent traffic in a statistical manner, though not as conclusive as with PCF.
FIG. 28 illustrates the manner in which preferential transmission is provided to traffic regarding which EDCF operations guarantee band. In the example shown in the drawings, a case is assumed wherein the STA1 attempts to transmit preferential traffic to the STA0, and the STA2 attempts to transmit non-preferential traffic to the STA0. Also, the standard IFS for both traffics is assumed to be time equivalent to DIFS.
Upon the media becoming clear from point-in-time T0, the STA1 and the STA2 both wait for the time of DIFS to elapse. The media is still clear following DIFS elapsing from T0 (point-in-time T1), so the STA1 and the STA2 both start to confirm that the media is clear at a time determined by random backoff.
With EDCF operations, the backoff value of the STA1 is short for preferential traffic, and the backoff value of the STA2 is long for non-preferential traffic. FIG. 28 illustrates the backoff values from point-in-time T1 of each of the communication stations with arrows. At point-in-time T2 where time of the backoff value of the STA1 has elapsed, the STA1 starts transmission of the RTS. On the other hand, the STA2 detects the RTS transmitted from the STA1, updates the backoff value, and prepares for the subsequent transmission.
Also, the STA0 returns a CTS at point-in-time T3 where SIFS has elapsed from reception of the RTS. The STA1 which has received the CTS starts data transmission at a point-in-time T4 where SIFS has elapsed from reception of CTS. The STA0 then returns an ACK at a point-in-time T5 where SIFS has elapsed from reception of data from the STA1.
At point-in-time T6 where returning of the ACK by the STA0 ends, the media is clear again. The STA1 and the STA2 both await elapsing of time of DIFS again. In the event that the media is still clear following elapsing of DIFS (point-in-time T7), the STA1 and the STA2 both start to confirm that the media is clear at a time determined by random backoff. Here also, the backoff value of the STA1 is set short due to preferential traffic, and RTS transmission is performed before the backoff value of the STA2 at point-in-time T8.
Due to the above-described procedures, order of access rights is provided to the STA1 and the STA2 competing for the access right, according to the degree of preference of the traffic being handled. Also, though not shown in the drawing, the backoff value of the STA2 also gradually becomes shorter, so a situation wherein the STA2 never gets access rights does not occur.